A Smart Property Management Method Based on the Internet of Things

Abstract

This invention discloses a smart property work order data management method based on the Internet of Things (IoT), relating to the field of smart community management technology. The method includes: matching fingerprinted event packets with a smart community work order database; employing multi-stage matching and filtering to construct a structured work order dataset; aggregating the dataset along spatiotemporal dimensions to form a structured evidence chain; and writing the chain into a community work order data lake after integrity verification. A quality control process is triggered based on verification tags to detect field missing rates, fingerprint duplication rates, and event temporal consistency, generating work order data packets and registering them in the smart community property big data lake repository. This invention uses a multi-dimensional spatiotemporal semantic verification algorithm to jointly assess spatial topological consistency and event causal semantic coherence, generating quantifiable quality verification tags. This ensures the structural integrity and semantic reliability of work order data before it is written into the big data lake repository, providing high-fidelity data support for the automated operation and maintenance of smart properties.

Description

Technical Field

[0001] This invention relates to the field of smart community management technology, and in particular to a smart property work order data management method based on the Internet of Things. Background Technology

[0002] In recent years, with the widespread adoption of IoT sensing in smart communities, property management work orders have gradually evolved from manual reporting to event-driven automated generation. Smart property management has initially achieved the collection and structured processing of raw sensor events, and is attempting to link historical work order data through spatiotemporal tags to support intelligent work order dispatch and traceability. By adopting semantic annotation and a data lake architecture, the manageability and analytical capabilities of work order data are being improved.

[0003] However, existing methods lack consistency verification between the intrinsic semantics of events and the external work order context during the transformation of multi-source heterogeneous event streams into high-quality work order data. There are gaps in spatial topology matching and event causal coherence, resulting in semantic drift, spatiotemporal misalignment and duplication of the generated work order data, which makes it difficult to support a high-reliability decision-making and service closed loop. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a smart property work order data management method based on the Internet of Things to solve the problem of spatial lack of consistency verification and continuity discontinuity in the Internet of Things event flow.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides a smart property work order data management method based on the Internet of Things (IoT). The method includes: collecting raw event streams from the IoT and generating standardized event packages through structured parsing; performing multi-dimensional feature coupling analysis on the standardized event packages to extract event features in spatial, temporal, and behavioral dimensions, generating an event fingerprint chain with identifiers and semantic tags, and encapsulating it into fingerprinted event packages; matching the fingerprinted event packages against a smart community work order database, employing multi-stage matching and filtering to construct a structured work order dataset, aggregating it according to spatiotemporal dimensions to form a structured evidence chain, and writing it into the community work order data lake after integrity verification; based on the structured evidence chain and event fingerprint chain, using a multi-dimensional spatiotemporal semantic verification algorithm to assess the confidence level of spatial topological consistency and event causal semantic coherence, and outputting quality verification labels; triggering a quality control process based on the verification labels to detect field missing rate, fingerprint duplication rate, and event temporal consistency, generating work order data packages, and registering them in the smart community property big data lake.

[0007] As a preferred embodiment of the IoT-based smart property work order data management method of the present invention, the original event stream refers to an unstructured event sequence output by IoT sensing nodes, containing timestamps, location identifiers and action descriptions, with heterogeneous data formats and no unified semantic annotation.

[0008] As a preferred embodiment of the IoT-based smart property work order data management method of the present invention, the specific steps for generating standardized event packages are as follows: Collect the raw event stream continuously output by the IoT sensing layer, and decompile the heterogeneous data frames using the payload decoding method to obtain intermediate event records in field format; The intermediate event records are semantically normalized and mapped to timestamps, source identifiers, and action types. Missing dimensions are filled with timestamps, location identifiers, and action descriptions to generate standardized event packages.

[0009] As a preferred embodiment of the IoT-based smart property work order data management method of the present invention, the encapsulation is a fingerprint-based event package, and the specific steps are as follows. Based on the geographic coordinates of standardized event packets, a dynamic neighborhood aggregation algorithm is used to jointly perceive and process spatial distribution and temporal rhythm to obtain preliminary feature clusters. Semantic parsing is performed on the action sequences in the preliminary feature clusters, and high-frequency behavior sequences and abnormal behavior fragments are extracted to generate an event fingerprint chain with identifiers and scene semantic labels. Through multidimensional feature coupling analysis, the event fingerprint chain with identifiers and scene semantic labels is coupled, encapsulated, and its integrity is verified, and fingerprinted event packets are output.

[0010] As a preferred embodiment of the IoT-based smart property work order data management method of the present invention, the smart community work order database refers to a data set that stores historical work order records, including work order semantic tags, service occurrence location, acceptance time window, and processing status information.

[0011] As a preferred embodiment of the IoT-based smart property work order data management method of the present invention, the community work order data lake is a centralized data storage area that stores and manages work order data records from the smart community.

[0012] As a preferred embodiment of the IoT-based smart property management work order data management method of the present invention, the specific steps of writing the data into the community work order data lake are as follows: Based on the semantic identifiers and spatiotemporal anchors of fingerprinted event packets, the semantic tag matching items, spatial topological proximity items, and time window overlap items of the smart community work order database are retrieved synchronously. Intersection constraints and co-occurrence consistency checks are performed to obtain a preliminary matching result set. Based on the semantic overlap between the event fingerprint chain in the preliminary matching results set and the smart community work order database, a multi-stage screening process is adopted to refine and filter candidate items, generating a structured work order dataset. The structured work order dataset is aggregated in spatiotemporal dimensions according to location proximity and temporal continuity to obtain a structured evidence chain, which is then written into the community work order data lake after integrity verification.

[0013] As a preferred embodiment of the IoT-based smart property work order data management method of the present invention, the specific steps for outputting the quality verification label are as follows: Based on the structured evidence chain and event fingerprint chain, a multi-dimensional spatiotemporal semantic verification algorithm is used to compare the consistency of spatial topological characteristics and obtain the matching degree between regional location sequence and behavioral trajectory. Based on the matching degree and the chronological order of events, the semantic coherence between actions is dynamically verified, and event chain confidence level labels are generated. Perform quality grading judgment on the confidence level label and output a quality verification label with identification.

[0014] As a preferred embodiment of the IoT-based smart property work order data management method of the present invention, the quality control process refers to the process of detecting and correcting the field integrity, fingerprint uniqueness, and time sequence correctness of work order data according to the confidence level indicated by the quality verification label.

[0015] As a preferred embodiment of the IoT-based smart property work order data management method of the present invention, the specific steps of writing the data into the smart community property big data warehouse are as follows: Based on the confidence level indicated by the quality verification label, the fingerprinted event package is subjected to field integrity scanning and repeatability comparison to obtain work order data indicators such as field missing ratio, fingerprint collision frequency and time series offset. Based on the work order data indicators, the structured work order dataset is filtered and reorganized through multi-stage screening to generate a work order data package; By combining the semantic tags of the event fingerprint chain and the spatiotemporal consistency information of the structured evidence chain, a semantic spatiotemporal joint consistency verification is performed on the work order data packet, and the work order data packet that passes the verification is written into the smart community property big data warehouse.

[0016] The beneficial effects of this invention are as follows: by using a multi-dimensional spatiotemporal semantic verification algorithm to jointly assess the spatial topological consistency and the semantic coherence of event causality, a quantifiable quality verification label is generated, ensuring the structural integrity and semantic reliability of work order data before it is written into the big data lake warehouse, and providing authentic data support for the automated operation and maintenance of smart properties. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a smart property management work order data management method based on the Internet of Things.

[0019] Figure 2 This is a flowchart illustrating the process of processing raw events into a structured chain of evidence in the Internet of Things (IoT) context.

[0020] Figure 3 This is a flowchart for data quality assessment using a multidimensional spatiotemporal semantic verification algorithm.

[0021] Figure 4 This is a flowchart for quality control and work order data entry verification. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 This is one embodiment of the present invention, which provides a smart property work order data management method based on the Internet of Things, including the following steps: S1: Collect raw event streams from the Internet of Things and generate standardized event packages through structured parsing.

[0026] S1.1: Raw event stream refers to an unstructured sequence of events output by IoT sensing nodes, containing timestamps, location identifiers, and action descriptions. These events have heterogeneous data formats and are not subject to unified semantic annotation.

[0027] The raw event stream continuously output by the IoT sensing layer is collected, and the heterogeneous data frames are destructured and depacked using the payload decoding method to obtain intermediate event records in field format.

[0028] Furthermore, when collecting unstructured event sequences containing timestamps, location identifiers, and action descriptions continuously output by IoT sensing nodes, the data is parsed frame by frame according to the payload decoding method of each sensing node. The data is then parsed to extract the business data area that actually carries timestamps, location identifiers, and action descriptions after removing the frame header, frame tail, and checksum transmission information from the original event stream. The parsed data content is then decomposed into identifiable time fields, source fields, and behavior fields according to the semantic roles of the fields (time, source, and behavior), forming an intermediate event record with a unified field arrangement order.

[0029] Specifically, the payload decoding method is the process of extracting the payload portion (i.e., the business data after removing the frame header and checksum transmission information) from the received raw event stream, and parsing it into identifiable structured fields (such as timestamps, device identifiers, and action codes) based on the payload encoding. It is a process of converting heterogeneous, unstructured raw byte streams into intermediate data records with semantic roles.

[0030] For example, data frame headers and checksums are removed to ensure data integrity. Timestamps, location identifiers, and action descriptions are extracted from the remaining data payload. Each field is parsed and transformed. Timestamps are standardized into a uniform format, location identifiers are converted into standard spatial identifiers, and action descriptions are mapped through a semantic dictionary to generate structured data records.

[0031] S1.2: Perform semantic normalization mapping of timestamps, source identifiers and action types on intermediate event records, and fill in missing dimensions with timestamps, location identifiers and action descriptions to generate standardized event packages.

[0032] Furthermore, based on the timestamps, location identifiers, and action descriptions contained in the unstructured event sequences output by IoT sensing nodes, the timestamps in intermediate event records are uniformly converted into time representations (hours, minutes, and seconds). The source identifiers in intermediate event records are mapped to spatial identifiers according to location encoding. The action types in intermediate event records are mapped to action semantic tags according to the action semantic dictionary. For missing timestamps, source identifiers, and action types in intermediate event records, time completion is performed by event records with the same source identifier within the adjacent time window, forming a standardized event package with complete fields and consistent semantics.

[0033] Specifically, mapping refers to the process of converting the timestamps, source identifiers, and action types in intermediate event records into a unified expression.

[0034] The mapping process involves parsing and reorganizing the timestamps of intermediate events into a unified time representation consisting of hours, minutes, and seconds. For the source identifier, it is mapped to a unified spatial identifier based on the location code, ensuring comparability of location descriptions across different IoT sensing nodes. For the action type, the original action description is replaced with a standardized action semantic label by matching terms from the action semantic dictionary through a lookup table.

[0035] Location identifiers, also known as source identifiers, refer to the physical location codes of IoT sensing nodes, identifying the geographical location where an event occurred.

[0036] S2: Perform multidimensional feature coupling analysis on the standardized event package, extract event features in spatial, temporal and behavioral dimensions, generate an event fingerprint chain with identifiers and semantic tags, and encapsulate it into a fingerprinted event package.

[0037] S2.1: Based on the geographic coordinates of standardized event packets, a dynamic neighborhood aggregation algorithm is used to jointly perceive and process spatial distribution and temporal rhythm to obtain preliminary feature clusters.

[0038] Furthermore, using the geographic coordinates contained in the standardized event package as input, neighborhood locations are constructed in space based on the spatial distribution of event locations. Combined with the timestamp information in the standardized event package, a sliding window is used to align the timing of events in the time dimension. Through a dynamic neighborhood aggregation algorithm, standardized event packages that are in the same or adjacent spatial neighborhoods and have the same timing are clustered into the same group to form an event set, which is the preliminary feature cluster.

[0039] It should be noted that the dynamic neighborhood aggregation algorithm is a joint sensing method based on the spatial distribution and temporal rhythm of events. It dynamically identifies the adjacency relationships between events by analyzing the geographical coordinates and timestamp information of events.

[0040] In the spatial dimension, by analyzing the geographical coordinates of events, spatially adjacent and identical events are grouped together. In the temporal dimension, based on the timestamp information of events, the algorithm dynamically adjusts the neighborhood range to determine events that are close in time and form adjacency relationships in the time series. Through dual perception of space and time, the clustering process of events is optimized, and events in similar spatial locations and time points can be identified and classified, providing event sets for feature extraction and data processing.

[0041] For example, adjacent access control points in the same building, unit, and corridor are used as the initial spatial neighborhood. At the same time, the time alignment scale is determined by the reporting cycle of the sensing nodes and the common trigger interval of historical events. For example, events that occur consecutively within the same alarm cycle are used as the basis for determining the same rhythm. During the aggregation process, if it is found that the number of events in the same spatial neighborhood is sparse, the interval between events is significantly longer, and the clustering information is divided into multiple fragments, the neighborhood range is broadened and the interrelated events are merged.

[0042] Conversely, if the aggregated events are found to span buildings, units, and significantly discontinuous time periods, resulting in spatial jumps and temporal breaks within the same cluster, the neighborhood scope is tightened to exclude irrelevant events.

[0043] S2.2: Perform semantic parsing on the action sequences in the preliminary feature clusters, extract high-frequency behavior sequences and abnormal behavior fragments, and generate an event fingerprint chain with identifiers and scene semantic labels.

[0044] Furthermore, based on the normalized action types in the standardized event package, semantic mapping is performed on the action sequences arranged in chronological order within the preliminary feature cluster. The action descriptions are converted into comparable labels in the semantic space. The frequency of occurrence of semantic labels within the time window is counted, and actions that recur and meet the continuity condition are identified as high-frequency behavior sequences. By comparing the action sequences with the regular behaviors in historical work order records, deviating behavior segments are detected as abnormal behavior segments. High-frequency behavior sequences and abnormal behavior segments are assigned identifiers respectively, and combined with the occurrence location and time context, scene semantic labels are added to form an event fingerprint chain composed of identifiers, scene semantic labels, and behavior sequence representations.

[0045] Specifically, meeting the continuity condition means that in an event sequence, events should maintain consistency in time, space, and behavior dimensions. There should be no abnormal jumps in the time intervals and spatial locations of events, and the sequence of behaviors should be orderly, avoiding unreasonable time, location, and behavioral intervals to ensure rationality and coherence. Historical work order records refer to the collection of service work order data stored in smart community property management, generated by past resident repair requests, equipment alarms, inspection tasks, and automatic event triggers. Each record includes work order semantic tags, service location, acceptance time window, and processing status information.

[0046] S2.3: Through multi-dimensional feature coupling analysis, the event fingerprint chain with identifiers and scene semantic labels is coupled, encapsulated, and its integrity is verified, and fingerprinted event packages are output.

[0047] Furthermore, based on the event characteristics of spatial, temporal, and behavioral dimensions, a multidimensional feature coupling analysis is performed on the event fingerprint chain with identifiers and scene semantic tags. The features of each dimension are aligned and fused under a unified spatiotemporal benchmark to form a coupled structured representation. The field integrity of the coupled structured representation is verified to confirm the existence of identifiers, the non-emptiness of scene semantic tags, and the consistency of timestamps and location identifiers, and the fingerprinted event package is output.

[0048] Specifically, multidimensional feature coupling analysis is the process of jointly aligning and fusing spatial features (such as geographic coordinates and regional identifiers), temporal features (such as the timing and rhythm of event occurrences), and behavioral features (such as high-frequency behavioral sequences and abnormal behavioral fragments) contained in the event fingerprint chain.

[0049] Under the spatiotemporal reference, the spatial, temporal and behavioral features in the event fingerprint chain are correlated and fused. Through the mutual constraints and enhancements of spatial proximity (such as multiple events occurring in the same building and adjacent areas), temporal continuity and behavioral semantic consistency (such as the behavioral chain composed of access control failure, abnormal face, and infrared movement), isolated and contradictory event fragments are eliminated, and a clear multidimensional event representation is generated.

[0050] For example, if Unit 2 of Building 3 experiences consecutive access control card swiping failures, facial recognition anomalies, and corridor infrared motion alarms within 10 seconds, and these events are adjacent in location, sequential in time, and semantically related in behavior, they will be coupled into a high-confidence representation of a suspected illegal intrusion event.

[0051] S3: Based on the fingerprinted event package, perform matching with the smart community work order database, adopt multi-stage matching and filtering to form a structured work order dataset, and aggregate it according to the spatiotemporal dimension to form a structured evidence chain. After integrity verification, write it into the community work order data lake.

[0052] S3.1: The smart community work order database refers to a data set that stores historical work order records, including work order semantic tags, service location, acceptance time window, and processing status information.

[0053] The community work order data lake is a centralized data storage area that stores and manages work order data records from smart communities.

[0054] Based on the semantic identifiers and spatiotemporal anchors of fingerprinted event packets, the semantic tag matching items, spatial topological proximity items, and time window overlap items are synchronously retrieved from the smart community work order database. Intersection constraints and co-occurrence consistency checks are performed to obtain a preliminary matching result set.

[0055] Furthermore, based on the semantic identifier of the fingerprinted event package, historical work order records with the same or equivalent semantic tags are searched in the smart community work order database. At the same time, based on the geographical coordinate information in the fingerprinted event package, the proximity between the service location and the location of each work order in the smart community work order database is calculated using spatial topology proximity terms. Work order records that meet the spatial proximity condition are filtered out. In addition, combined with the timestamp information of the fingerprinted event package, work order records in the smart community work order database that have overlapping acceptance time windows and timestamps are searched.

[0056] The semantic tag matching items, spatial topological proximity items, and temporal window overlap items of the retrieved information are subjected to set intersection operation. The work order records that simultaneously meet all three conditions are retained as candidate sets. The co-occurrence stability of each work order record in the candidate set with the fingerprinted event package in the semantic, spatial and temporal dimensions is checked for consistency. That is, it is verified that the multidimensional feature coupling analysis has behavioral semantic coherence and spatiotemporal co-occurrence relationship (meaning that the event and the work order simultaneously meet the matching, proximity and overlap in semantic tag, spatial location and time window). The preliminary matching result set is output.

[0057] Specifically, an event fingerprint chain is a sequence structure composed of event fingerprints arranged in chronological order.

[0058] An event fingerprint is the fingerprint content in an event fingerprint chain, which includes an identifier, a scene semantic label, and a behavioral sequence representation.

[0059] A fingerprinted event package is a data object that encapsulates a standardized event package and its corresponding event fingerprint chain.

[0060] The preliminary matching result set is based on the semantic identifiers and spatiotemporal anchors of fingerprinted event packets. Historical work order records that simultaneously meet the three conditions of semantic tag matching, spatial topological proximity, and time window overlap are retrieved from the smart community work order database. After intersection constraint and co-occurrence consistency verification, the candidate work order set is obtained.

[0061] S3.2: Based on the semantic overlap between the event fingerprint chain in the preliminary matching result set and the smart community work order database, a multi-stage screening process is adopted to refine and filter the candidate items, generating a structured work order dataset.

[0062] Furthermore, based on the scene semantic tags in the event fingerprint chain and the work order semantic tags of each record in the smart community work order database, a multi-stage screening is adopted to retain candidate entries with co-occurrence relationships in semantic tags. Combining spatial topological proximity items and time window overlap items, the intersection constraint verification is performed on the retained candidate entries to eliminate entries with spatial location deviation and no time window overlap. Based on the temporal consistency between the behavior sequence in the event fingerprint chain and the service occurrence location and acceptance time window in the smart community work order database, the remaining candidate entries are refined and filtered, and the candidate entries that have passed the multi-stage screening are integrated into a structured work order dataset.

[0063] It should be noted that the refinement filtering refers to: semantic layer filtering, which removes entries where the semantic tags of the event scenario do not match or are not equivalent to the semantic tags of the work order; spatiotemporal layer filtering, which removes entries where the service location is beyond spatial proximity and the acceptance time window does not overlap with the event time; and behavioral layer filtering, which removes entries where the historical processing action sequence and the current event behavior sequence are not consistent in terms of time and semantics.

[0064] Multi-stage screening is an orderly and progressive filtering process performed on the initially matched candidate items during the smart community work order matching process. It is based on the semantic tags of events and work orders for matching and screening, and combines spatial proximity and time window overlap to perform spatiotemporal consistency verification, verifying the temporal and semantic coherence between the event behavior sequence and the historical actions of the work orders.

[0065] Each stage only passes entries that meet the current conditions to the next stage. By eliminating inconsistent and low-confidence records layer by layer, a structured single-item dataset with semantic accuracy, spatiotemporal alignment, and reasonable behavior is generated.

[0066] Semantic overlap refers to the degree of matching and equivalence between the scene semantic tags in the event fingerprint chain and the work order semantic tags in the historical work order records of the smart community work order database at the business semantic level. It measures the description of the same type of event and has a reasonable causal relationship (such as access control failure and security verification).

[0067] S3.3: Aggregate the structured work order dataset according to the spatiotemporal dimensions of location proximity and time continuity to obtain a structured evidence chain, and write it into the community work order data lake after integrity verification.

[0068] Furthermore, based on the service location and acceptance time window contained in each record of the structured work order dataset, work order records that are spatially adjacent and have consecutive and overlapping acceptance time windows are grouped into the same spatiotemporal cluster according to the proximity of geographical coordinates. Within each spatiotemporal cluster, work order records are arranged in chronological order to form an event sequence with temporal and spatial correlation. The event sequence is subjected to integrity verification, which includes the geographical identifier of the service location, the temporal continuity of the acceptance time window, and missing items of processing status information. When the integrity verification passes, the event sequence is used as a structured evidence chain, aggregated through the spatiotemporal dimension, retaining the corresponding content of the original event and work order record, and written into the community work order data lake.

[0069] The formula for the completeness of a structured chain of evidence is: ; in, It is the integrity score of the structured chain of evidence, dimensionless (1 means all records are complete, 0 means all are missing). It is the number of work order records contained in the structured evidence chain (dimensionless, representing a count). It's an index of work order records, a dimensionless integer index. It is the first The integrity indicator value of each work order record (dimensionless and a constant).

[0070] It should be noted that all variables ( , ,r, , All of these are dimensionless quantities. Indicator values ​​are 0 and 1. For counting, the result of the operation For the sake of proportion, dimensionless, and the dimensional unification of the formula for the integrity of the structured chain of evidence.

[0071] Spatiotemporal aggregation is a process of grouping work order records in a structured work order dataset that are located in adjacent locations (such as the same building, the same group, or within the same geographical radius) and whose acceptance time windows are consecutive and overlapping into the same spatiotemporal cluster. Within the same spatiotemporal cluster, the correspondence between events and work orders is organized in chronological order, forming an aggregation combination with spatial regions and time intervals as the granularity.

[0072] For example, access control failure events and related work orders that occurred in Unit 2 of Building 3 between 08:15 and 08:30, and facial recognition anomaly events and work orders between 08:16 and 08:25, are aggregated into a spatiotemporal cluster identified by Unit 2 of Building 3 between 08:15 and 08:30 due to their proximity and overlapping time. Meanwhile, lighting faults in Building 5 are formed into independent clusters due to spatial separation. Each spatiotemporal cluster is written into the community work order data lake as a structured unit, which not only preserves the complete correspondence between the original events and work orders, but also realizes data aggregation organized according to the real physical scene and time context.

[0073] S4: Based on the structured evidence chain and event fingerprint chain, a multi-dimensional spatiotemporal semantic verification algorithm is used to evaluate the confidence of spatial topological consistency and event causal semantic coherence, and output quality verification labels.

[0074] S4.1: Based on the structured evidence chain and event fingerprint chain, a multi-dimensional spatiotemporal semantic verification algorithm is used to compare the consistency of spatial topological characteristics and obtain the matching degree between regional location sequence and behavioral trajectory.

[0075] Furthermore, the spatiotemporal anchor points formed by aggregating the structured evidence chain according to location proximity and temporal continuity are mapped point by point with the geographic coordinates and temporal rhythm joint perception results generated by the dynamic neighborhood aggregation algorithm in the event fingerprint chain. Based on the spatial topology, the spatial adjacency of the regional location sequence represented by the structured evidence chain is consistent with the behavioral trajectory path described by the event fingerprint chain. Combined with the chronological order of event occurrence, the spatial evolution direction of the regional location sequence and the temporal advancement direction of the behavioral trajectory are verified. The consistency between spatial adjacency overlap and temporal evolution direction is quantified by a multi-dimensional spatiotemporal semantic verification algorithm, and the matching degree between the regional location sequence and the behavioral trajectory is output.

[0076] The formula for calculating the matching degree between regional location sequences and behavioral trajectories is as follows: ; in, It is the first The degree of matching between individual behavioral trajectories and regional location sequences. It is the sequence number of the behavioral trajectory. It is the number of matched region location sequences. It is the first The location sequence of the region and the first The matching degree of a behavioral trajectory is the matching degree obtained by aligning the region location sequence formed by aggregating location proximity and temporal continuity in the structured evidence chain with the behavioral trajectory generated by the dynamic neighborhood aggregation algorithm in the event fingerprint chain point by point.

[0077] It should be noted that, The matching score is a unitless score, usually a value between 0 and 1, and is dimensionless. It simply represents the number of the behavioral trajectory, serving as a sequence number; it has no unit and is dimensionless. This represents the number of regional location sequences that are matched with the behavioral trajectory. The number itself is unitless and dimensionless. Matching degree represents the consistency between a region's location sequence and a behavioral trajectory, and is usually expressed as a dimensionless value between 0 and 1.

[0078] The multidimensional spatiotemporal semantic verification algorithm is a driver-based joint verification method that verifies the consistency of the event fingerprint chain and the structured evidence chain in space, time, and semantics.

[0079] By verifying that the sequence of event occurrences and the location of the work order service are adjacent and connected in the community spatial topology, that the event timestamp is earlier than the work order acceptance time window, and that the service response content is composed of the event behavior type (such as access control failure) and the work order semantic tag (such as security verification), when all of these conditions are met, the data is identified as having high confidence and a quality verification tag is generated to ensure that the work order data written to the data lake has spatiotemporal rationality and business semantic credibility.

[0080] S4.2: Based on the matching degree and the chronological order of events, dynamically verify the semantic coherence between actions and generate event chain confidence level labels.

[0081] Furthermore, based on the timestamp sequence of each action in the event fingerprint chain, the actions in the event fingerprint chain are sorted by timestamp from smallest to largest to form an ordered sequence reflecting the chronological order of the actions. Combined with the service occurrence location and acceptance time window of the corresponding work order record in the structured evidence chain, the sequential logic of the action sequence in the time dimension is verified to see if it conforms to the conventional process of community service response. The scene semantic tags in the event fingerprint chain are compared with the work order semantic tags in the structured evidence chain to analyze the semantic relationship between adjacent actions. A multi-dimensional spatiotemporal semantic verification algorithm is used to jointly evaluate the spatial topological consistency and the causal semantic coherence of the event. If the action sequence has no inverted conflict in time order and has contextual correlation in semantic tags, a high event chain confidence level is assigned, and an event chain confidence level tag is generated.

[0082] It should be noted that conforming to the standard community service response process means, for example, that after a card swipe failure at an access control system, if subsequent actions such as facial recognition anomaly, security personnel arrival, and access control device restart are detected in sequence, and the corresponding work order records are security verification and equipment maintenance, then the sequence conforms to the common community security alarm, manual confirmation, and technical handling response process. Conversely, if the incident is a lighting malfunction, but the work order action is to replace the door lock before reporting the electrician for repair, then the timing and business logic are disordered and do not conform to the standard process.

[0083] The event chain confidence level is a confidence level assigned to the sequence of events and work orders associated with the event fingerprint chain and the structured evidence chain. It is based on the spatial proximity and temporal order (events occurring before work orders are accepted) and the consistency of behavioral semantics (e.g., access control anomalies corresponding to security checks). High confidence indicates consistency and compliance with the regular community service process; medium confidence indicates slight deviations in some dimensions but still acceptable; and low confidence indicates spatiotemporal misalignment and semantic contradictions. As a quality verification result, it determines whether the data is directly written to the lake warehouse and whether correction and review are required.

[0084] Spatial topological consistency means that the sequence of locations where events occur and the corresponding work order service locations are adjacent in the community's physical spatial structure (such as buildings, units, and road connections), and are consistent in spatial layout, without cross-regional jumps or geographical contradictions.

[0085] Event causal semantic coherence means that the sequence of behaviors in the event fingerprint chain and the work order actions in the structured evidence chain constitute a reasonable causal and response relationship in semantics (such as access control anomaly, security check and equipment maintenance). The order of events conforms to the business order of problem triggering, response and handling in community services, rather than a random and inverted semantic combination.

[0086] S4.3: Perform quality grading judgment on the confidence level label and output a quality verification label with identification.

[0087] Furthermore, based on the spatial topological consistency comparison content and the dynamic verification content of semantic coherence of behavior and action reflected by the event chain confidence level label, the event chain confidence level label is divided into high, medium and low quality levels according to the classification of the multi-dimensional spatiotemporal semantic verification algorithm, and a corresponding identifier is attached to each level to form a quality verification label with label.

[0088] Specifically, the quality grading determination involves dividing work order data into three quality levels: high, medium, and low, and attaching a clear identifier (ABC) to each level to form a quality verification label with the identifier.

[0089] For example, when the matching degree between the regional location sequence and the behavior trajectory is ≥0.9 and the timing of the behavior actions conforms to the regular community service process (such as access control failure → security personnel arriving → device restart), it is judged as high confidence, and the A label is output, allowing direct writing to the big data lake warehouse. If there is spatiotemporal offset and weak semantic association, it is marked as B and field completion and review are triggered. If there is cross-building jump, time inversion and semantic contradiction, it is marked as C and intercepted for storage.

[0090] S5: Trigger the quality control process based on the verification tag, detect the field missing rate, fingerprint duplication rate and event sequence consistency, generate work order data package, and register and write it into the smart community property big data warehouse.

[0091] S5.1: The quality control process refers to the process of detecting and correcting the field integrity, fingerprint uniqueness, and time sequence correctness of work order data based on the confidence level indicated by the quality verification label.

[0092] Based on the confidence level indicated by the quality verification label, the fingerprinted event package is subjected to field integrity scanning and repeatability comparison to obtain work order data indicators such as field missing ratio, fingerprint collision frequency and time series offset.

[0093] Furthermore, based on the confidence level indicated by the quality verification label, each field in the fingerprinted event package is traversed, and the ratio of the number of unfilled and null fields to the total number of fields is calculated to obtain the field missing ratio. The event fingerprint chain in the current fingerprinted event package is compared item by item with the event fingerprint chain already stored in the community work order data lake, and the number of times the same identifier and scene semantic label combination appears is recorded to obtain the fingerprint collision frequency. The timestamp sequence of the event in the fingerprinted event package is extracted and aligned sequentially with the time sequence of the corresponding event in the structured evidence chain to obtain the cumulative offset difference of the time interval between adjacent events, forming the work order data index of time series offset.

[0094] S5.2: Based on the work order data indicators, the structured work order dataset is filtered and reorganized through multi-stage screening to generate a work order data package.

[0095] Furthermore, based on the field missing ratio, integrity filtering is performed on each entry in the structured work order dataset to remove records with a field missing ratio exceeding the work order ratio. Fingerprint verification is performed on the retained records based on the fingerprint collision frequency to remove redundant entries that repeatedly appear in the event fingerprint chain. The timestamp continuity of the remaining records is verified based on the time series offset to exclude event records with time sequence disorder and jump abnormalities. The structured work order dataset that has passed the screening at each stage is reorganized according to semantic label consistency and spatiotemporal anchor proximity to form a work order data package that meets the quality requirements.

[0096] Specifically, meeting the quality requirements means that after the work order data package has undergone field integrity, fingerprint uniqueness, and time sequence continuity checks, it has complete necessary fields, a non-duplicate event fingerprint chain, and a coherent and error-free timestamp sequence. Furthermore, the semantic tags and spatiotemporal anchors are consistent with each other, and it can truly and reliably reflect the correspondence between community events and work orders.

[0097] S5.3: Combining the semantic tags of the event fingerprint chain and the spatiotemporal consistency information of the structured evidence chain, perform semantic spatiotemporal joint consistency verification on the work order data packet, and write the work order data packet that passes the verification into the smart community property big data warehouse.

[0098] Furthermore, based on the scene semantic tags contained in the event fingerprint chain, they are compared with the spatiotemporal anchor points formed by aggregation according to location proximity and temporal continuity in the structured evidence chain. The behavior type indicated by the semantic tags is verified to occur within the corresponding spatial topology region and time window through a multi-dimensional spatiotemporal semantic verification algorithm. It is also checked whether the action sequence in the event fingerprint chain and the event sequence in the structured evidence chain meet the sequential constraints. If both the spatial topology consistency and the event causal semantic coherence pass the confidence assessment, the work order data packet is determined, the consistency verification content is obtained, and the work order data packet with the consistency verification content is written into the smart community property big data warehouse.

[0099] Specifically, spatiotemporal anchor point comparison refers to matching the geographical coordinates and timestamps in the event fingerprint chain with the spatial region and time interval represented by the spatiotemporal anchor points, respectively. If the event location falls within the spatial region and the event time is within the time interval, it is considered a match.

[0100] Confidence assessment is a process of determining the authenticity and reliability of the association between an event and a work order based on the matching relationship between the event fingerprint chain and the structured evidence chain in terms of spatial topological consistency (whether the locations are adjacent and connected), temporal sequence rationality (whether the event precedes the work order and the order is in compliance with regulations), and behavioral semantic coherence (whether the action and the work order constitute a reasonable response relationship). It also generates high, medium, and low-level quality verification labels.

[0101] The consistency verification refers to the judgment conclusion obtained by jointly verifying the event fingerprint chain and structured evidence chain in the work order data packet in terms of semantic labels, spatial location, and time sequence. That is, to confirm whether the three conditions are met simultaneously: the behavior type is consistent with the semantics of the work order, the event location is adjacent to the service location, and the event time sequence is consistent with the time logic of the work order. If all three conditions are met, it is determined that the consistency verification has passed and can be written into the smart community property big data warehouse; otherwise, it is intercepted and marked as low-quality data.

[0102] A smart community property management big data lake warehouse is a centralized data storage area used to store and manage various data from the smart community, including event streams generated by the Internet of Things, work order records, sensor data, and behavior logs. Through unified storage and management, it integrates raw and structured data from different sources and performs integrity checks. Data that passes the checks is stored in the data lake warehouse, supporting data analysis, event matching, and intelligent decision-making. The data lake warehouse provides smart communities with powerful data query, analysis, and mining capabilities, supporting the intelligent management and optimization of the community.

[0103] In summary, this invention uses a multi-dimensional spatiotemporal semantic verification algorithm to jointly assess the spatial topological consistency and the semantic coherence of event causality, generating quantifiable quality verification labels to ensure the structural integrity and semantic reliability of work order data before it is written into the big data lake warehouse, thus providing authentic data support for the automated operation and maintenance of smart properties.

[0104] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A smart property management work order data management method based on the Internet of Things, characterized in that: include, Collect raw event streams from the Internet of Things and generate standardized event packages through structured parsing; Multidimensional feature coupling analysis is performed on standardized event packages to extract event features in spatial, temporal, and behavioral dimensions, generate event fingerprint chains with identifiers and semantic tags, and encapsulate them into fingerprinted event packages; The fingerprinted event package is matched against the smart community work order database. Multi-stage matching and filtering are used to form a structured work order dataset, which is then aggregated according to the spatiotemporal dimension to form a structured evidence chain. After integrity verification, the data is written into the community work order data lake. Based on the structured evidence chain and event fingerprint chain, a multi-dimensional spatiotemporal semantic verification algorithm is used to evaluate the confidence of spatial topological consistency and event causal semantic coherence, and output quality verification labels. The quality control process is triggered based on the verification label, which detects the field missing rate, fingerprint duplication rate and event sequence consistency, generates a work order data package and registers it in the smart community property big data warehouse.

2. The smart property management work order data management method based on the Internet of Things as described in claim 1, characterized in that: The raw event stream refers to an unstructured sequence of events output by IoT sensing nodes, containing timestamps, location identifiers, and action descriptions. These events have heterogeneous data formats and are not subject to unified semantic annotation.

3. The smart property management work order data management method based on the Internet of Things as described in claim 2, characterized in that: The specific steps for generating the standardized event package are as follows. Collect the raw event stream continuously output by the IoT sensing layer, and decompile the heterogeneous data frames using the payload decoding method to obtain intermediate event records in field format; The intermediate event records are semantically normalized and mapped to timestamps, source identifiers, and action types. Missing dimensions are filled with timestamps, location identifiers, and action descriptions to generate standardized event packages.

4. The smart property management work order data management method based on the Internet of Things as described in claim 3, characterized in that: The encapsulation is performed as a fingerprinted event packet, and the specific steps are as follows. Based on the geographic coordinates of standardized event packets, a dynamic neighborhood aggregation algorithm is used to jointly perceive and process spatial distribution and temporal rhythm to obtain preliminary feature clusters. Semantic parsing is performed on the action sequences in the preliminary feature clusters, and high-frequency behavior sequences and abnormal behavior fragments are extracted to generate an event fingerprint chain with identifiers and scene semantic labels. Through multidimensional feature coupling analysis, the event fingerprint chain with identifiers and scene semantic labels is coupled, encapsulated, and its integrity is verified, and fingerprinted event packets are output.

5. The smart property management work order data management method based on the Internet of Things as described in claim 4, characterized in that: The smart community work order database refers to a data set that stores historical work order records, including work order semantic tags, service location, acceptance time window, and processing status information.

6. The smart property management work order data management method based on the Internet of Things as described in claim 5, characterized in that: The community work order data lake is a centralized data storage area that stores and manages work order data records from smart communities.

7. The smart property management work order data management method based on the Internet of Things as described in claim 6, characterized in that: The structured evidence chain, formed by aggregation along the spatiotemporal dimension, is written into the community work order data lake after integrity verification. The specific steps are as follows: Based on the semantic identifiers and spatiotemporal anchors of fingerprinted event packets, the semantic tag matching items, spatial topological proximity items, and time window overlap items of the smart community work order database are retrieved synchronously. Intersection constraints and co-occurrence consistency checks are performed to obtain a preliminary matching result set. Based on the semantic overlap between the event fingerprint chain in the preliminary matching results set and the smart community work order database, a multi-stage screening process is adopted to refine and filter candidate items, generating a structured work order dataset. The structured work order dataset is aggregated in spatiotemporal dimensions according to location proximity and temporal continuity to obtain a structured evidence chain, which is then written into the community work order data lake after integrity verification.

8. The smart property management work order data management method based on the Internet of Things as described in claim 7, characterized in that: The specific steps for setting the output quality verification label are as follows: Based on the structured evidence chain and event fingerprint chain, a multi-dimensional spatiotemporal semantic verification algorithm is used to compare the consistency of spatial topological characteristics and obtain the matching degree between regional location sequence and behavioral trajectory. Based on the matching degree and the chronological order of events, the semantic coherence between actions is dynamically verified, and event chain confidence level labels are generated. Perform quality grading judgment on the confidence level label and output a quality verification label with identification.

9. The smart property management work order data management method based on the Internet of Things as described in claim 8, characterized in that: The quality control process refers to the process of detecting and correcting the field integrity, fingerprint uniqueness, and time sequence correctness of work order data based on the confidence level indicated by the quality verification label.

10. The smart property management work order data management method based on the Internet of Things as described in claim 9, characterized in that: The specific steps for writing the data into the smart community property big data warehouse are as follows: Based on the confidence level indicated by the quality verification label, the fingerprinted event package is subjected to field integrity scanning and repeatability comparison to obtain work order data indicators such as field missing ratio, fingerprint collision frequency and time series offset. Based on the work order data indicators, the structured work order dataset is filtered and reorganized through multi-stage screening to generate a work order data package; By combining the semantic tags of the event fingerprint chain and the spatiotemporal consistency information of the structured evidence chain, a semantic spatiotemporal joint consistency verification is performed on the work order data packet, and the work order data packet that passes the verification is written into the smart community property big data warehouse.

Images